dotnet
What is the "W" in this RGBAW color space matrix
Type of issue
Missing information
Description
That article helped me a lot to massively speed up an image invert and color change in powershell. But what is the "W" in the color space here? I tried to find anything, but all search results link back to this article.
Page URL
https://learn.microsoft.com/en-us/dotnet/api/system.drawing.imaging.colormatrix?view=windowsdesktop-9.0
Content source URL
https://github.com/dotnet/dotnet-api-docs/blob/main/xml/System.Drawing.Imaging/ColorMatrix.xml
Document Version Independent Id
965b4c7a-083d-b594-f75d-e4f63b71e2e0
Article author
@dotnet-bot
White.
Thank you for your response in the 24th of December! I suspected too, but... How or when is it applied? I changed the fifth value to 0, 0.5, 1, 1000, -1000 in each row. And the outcome is no different, pixel exact the same transformation as the rest of that colormatrix says. Is it applied only when a specific colorspace is used? The source and destination picture it is applied to is: [System.Drawing.Imaging.PixelFormat]::Format24bppRgb. Or is there something else how this value is used?
The whole powershell code used is this, to invert a white background diagram into dark color scheme and then gamma up the lower values to be better visible as a non-intrusive regularly updated windows wallpaper. $bmp is the original to be transformed. (Code to update the currently shown wallpaper is left out)
# https://learn.microsoft.com/en-us/dotnet/api/system.drawing.imaging.colormatrix
$ColorMatrix = [System.Drawing.Imaging.ColorMatrix]::new( [float[][]] @(
[float[]](-0.2, 0, 0, 0, 0), # Invert R
[float[]](0, -0.2, 0, 0, 0), # Invert G
[float[]](0, 0, -0.2, 0, 0), # Invert B
[float[]](0, 0, 0, 1, 0), # Alpha: Do nothing
[float[]](0.2, 0.2, 0.2, 0, 1) # Translate/Offset RGBA
)
)
# https://learn.microsoft.com/en-us/dotnet/api/system.drawing.imaging.imageattributes
$ImageAttributes = [System.Drawing.Imaging.ImageAttributes]::new()
$ImageAttributes.SetColorMatrix($ColorMatrix)
# https://learn.microsoft.com/en-us/dotnet/api/system.drawing.imaging.imageattributes.setgamma
$ImageAttributes.SetGamma(0.6)
$bmpinv = [System.Drawing.Bitmap]::new($bmp.Width,$bmp.Height,[System.Drawing.Imaging.PixelFormat]::Format24bppRgb)
$pictinv = [System.Drawing.Graphics]::FromImage($bmpinv)
$pictinv.DrawImage($bmp, # Source image
[System.Drawing.Rectangle]::new(0,0,$bmp.Width,$bmp.Height), # Destination rectangle, here same as source
0,0,$bmp.Width,$bmp.Height, # Source rectangle.
[System.Drawing.GraphicsUnit]::Pixel, # We want 1:1 not inch or whatever
$ImageAttributes # Finally the matrix
)
$bmpinv.Save("$DataBaseDir\today-inv.png",[System.Drawing.Imaging.ImageFormat]::Png)
$pictinv.Dispose()
$bmpinv.Dispose()
Can't offer a lot here, other than this is GDI+ under the covers. You might find something useful here: https://learn.microsoft.com/windows/win32/gdiplus/-gdiplus-recoloring-use.
Can't offer a lot here, other than this is GDI+ under the covers. You might find something useful here: https://learn.microsoft.com/windows/win32/gdiplus/-gdiplus-recoloring-use.
Oh! That explains it. It is actually a 4x5 matrix, with the fifth row being the non-matrix translation (or offset, if you like). But since a 4x5 matrix in "unwanted" in math, and probably unwanted when using hardware acceleration, the fifth column is added but unused. The article states what the fifth column must always be 0/0/0/0/1 for consistency reasons, but either it gets completely ignored by the math inside by design (unused), gets deliberately ignored by the code (not read the column and use correct value) or somewhere it gets force-set to 0/0/0/0/1 (read that column and correct). This means the DOTNET-Article is wrong, or unclear. It is not RGBAW space it is RGBA[dummy] space. My suggestion would be to add the fourth paragraph to the DOTNET article and state that "W" is that placeholder to get a clean 5x5 matrix. So: What does the code behind it do? Not use, ignore, or enforce?